Hyperammonemia with or without liver disease results in swollen astrocytes and large increases in brain glutamine levels. Ammonia is metabolized by an active glutamine synthetase pool localized in astrocytes. We observed that clinically-relevant elevations of plasma ammonia levels (approximately 600 muM) in anesthetized rats cause increases in intracranial pressure and cortical water content associated with a tripling of tissue glutamine levels. Remarkably, we found that inhibiting glutamine synthetase activity with methionine sulfoximine (MSO) not only prevented the increase in tissue glutamine, but also prevented the increase in water content and intracranial pressure. Moreover, the cerebral blood flow (CBF) responses to CO2 were completely abnormal during hyperammonemia: CBF increased with hypocapnia and decreased with hypercapnia. These abnormal CBF responses were largely reversed by MSO pretreatment. We will investigate several testable hypotheses to account for cerebral edema, intracranial hypertension and abnormal cerebrovascular responses. First, cerebral edema may be attributed to the osmotic load of the large increase in tissue glutamine, particularly if the increase is confined to the small astrocyte compartment. Such swelling may be a reversible process. We will test if reducing glutamine by MSO posttreatment after edema is formed in hyperammonemic rats reverses edema and astrocyte swelling. Second, amelioration of intracranial hypertension by MSO may be due to inhibition of the known stimulation of CSF production during hyperammonemia. Third, we will evaluate microcirculatory compression by swollen astrocyte foot processes and determine if extraparenchymal pial arterioles not subjected to direct astrocyte compression retain CO2 reactivity when intraparenchymal CBF is abnormal. Fourth, using ion selective microelectrodes, we will determine if astrocyte regulation of extracellular K+ activity and pH with CO2 alterations is abnormal. Fifth, we will determine if ammonia or glutamine, which is capable of inhibiting endothelial dependent relaxation, has direct effects on pial arteriolar reactivity in vivo. Therefore, hyperammonemia represents a unique model for demonstrating interaction of astrocyte function and vascular reactivity, thereby providing new insights into the physiology of cerebrovascular control mechanisms. Moreover, abnormal cerebrovascular regulation and swollen astrocytes surrounding capillaries may cause additional tissue hypoxia, and account for many of the pathophysiological abnormalities found in a wide range of diseases associated with hyperammonemia.